車美仕 (KMX) 2004 Q2 法說會逐字稿

Good morning.

My name is Leigh and I will be your conference facilitator.

At this time I would like to welcome everyone to the CarMax second-quarter earnings conference call. (OPERATOR INSTRUCTIONS) Ms. Barrett, you may begin your conference.

Good morning.

This is Dandy Barrett, Assistant Vice President of Investor Relations for CarMax.

Thank you for joining us this morning.

On the call today are Austin Ligon, our President and Chief Executive Officer, and Keith Browning, our Executive Vice President and Chief Financial Officer.

Before we begin, please let me remind you that our statements today about the Company's future business plans, prospects, and financial performance are forward-looking statements that we make relying on the Safe Harbor provisions of the Private Securities Litigation Reform Act of 1995.

These statements are based on management's current knowledge and assumptions about future events.

They involve risks and uncertainties that could cause actual results to differ materially from our expectations.

For additional information on important factors that could affect these expectations, please see the Company's Annual Report on Form 10-K for the fiscal year ended February 29, 2004 and our Quarterly and current reports on file with the SEC.

Now I will turn the call over to Austin.

Good morning.

Thanks for joining us.

Let me start with sales.

As you saw, we had a 7 percent increase in total sales, a 4 percent increase in total used sales and a 7 percent decline in comp store used unit sales.

Beginning in April, we saw widespread market volatility and softness.

And as you know, this continued through the summer, all the way through August.

And this five-month period of volatility is really unprecedented in our history in terms of the impact it's had on us.

As usual, differences in our performance -- we have differences in performance market to market, and even within market by store to store.

But overall, we saw the softness spread throughout the country.

As far as earnings go, second-quarter earnings as a result of the sales results were 29.9 million or 28 cents a share versus 39.6 million or 37 cents last year.

The factors that we believe contributed to this a soft used car market start with the combination of the economy and competition.

We believe that the unusually intense and volatile competition among new car manufacturers, particularly the domestic brands, led to unpredictable incentive behavior throughout the peak selling season of the summer, and made it difficult for the wholesale market to be as efficient as normal in adjusting to changes in new car pricing.

We've also been seeing a lower mix of prime customers coming through the door, which is consistent with the idea that new cars are cannibalizing some of our better used car customers.

The adjustment of wholesale pricing and the difficulties there was exacerbated by the inventory buildup of the new cars in the spring in advance of an anticipated robust summer selling season, as well as the used car industry's optimism during this spring when most new car and used car independent dealers added inventory expecting a strong spring and summer.

There was also a near-term decrease, as noted by many, in the supply of off-lease vehicles.

It's difficult for us to predict when this market volatility will ease.

It's reasonable to expect that the market would rationalize somewhat as we move through the model year changeover period this fall.

However, our third-quarter forecast assumes a continuation of the soft used car market conditions.

Based on our analysis of the available market share information, we believe that we continue to maintain or gain market share.

We added during the summer, and really primarily in August, a new subprime -- true subprime financed provider, Drive Financial Services.

This had a modest impact on overall sales for the quarter given the timing of the actual rollout during August.

And it's a bit early for us to predict the ultimate incremental sales opportunity.

We do believe there's some opportunity for both incremental sales and profit, and roughly we're thinking that low-single-digit sales impact is the likely range of sales impact.

Also it's worth noting that Drive produces roughly 40 percent the net margin contribution per car as CarMax car does.

As is customary in the subprime finance industry, Drive purchases loan contracts at a discount, so we actually pay Drive when Drive does a financing.

And that's what drives the total contribution.

Plus we don't get the upside contribution that we would get from having a positive commission from one of our other suppliers.

As far as used car gross margins go, gross margin profit dollars per unit declined to 18.46 versus 18.60 last year.

These softer margins reflect the generally slower sales environment.

Also they reflect the fact that in June and July we maintained inventories somewhat higher than our run rates during those months because being peak sales season we wanted to make sure that in sales did in fact improve during the summer we had inventory to meet that increased demand given that it's very difficult to build inventory during the summer because of the high sales level.

Stronger sales did not materialize, obviously, which reduced our overall inventory turns and reduced our gross margins.

During August we reduced inventories to the target levels consistent with sales by the end of the month.

As you know, we always have a target of having our lowest inventory of the year during the week following Labor Day to minimize inventory risk during the fall period where you have both new car model changeover, the biggest fall in used car prices, and following demand sequentially as we go through the fall into December.

As far as CarMax Auto Finance goes, the CAF (ph) spread was 3.9 percent in the second quarter, near the midpoint of our normalized 3.5 to 4.5 percent range.

Last year the gain spread was 4.8 percent, reflecting some continued windfall last year.

For the balance of this year we expect the cap gain spread to be closer to the lower end of the normalized range -- as I said, closer to the 3.5 end of the normalized range.

The very slow pace of interest rate increases has made it more difficult to raise consumer rates, and we expect that to continue for the rest of the quarter.

The SG&A ratio at 10.2 percent was up 40 basis points from last year.

This increase was substantially all related to the deleveraging impact of negative used unit comps.

As far as third-quarter expectations, we assume that, as I said, market conditions continue to be volatile throughout the third quarter.

We are, however, encouraged by sales so far in September, even in spite of the hurricane pattern that we have had.

But it's too early to predict that this trend will hold, particularly during the always challenging fall with both new model changeover and seasonally falling demand.

As a result, we set third-quarter expectations consistent with the summer trend.

We are currently forecasting used unit comps of -2 to -8 percent, which would result in EPS in a range of 12 to 17 cents.

Given this level of uncertainty about the market trend and market dynamics we're not providing a fourth-quarter estimate at this time, because really we don't have a trend that would let us confidently forecast where that would be, either good or bad.

So with that, I'd be happy to take your questions.

(OPERATOR INSTRUCTIONS) Aram Rubinson, Banc of America Securities.

Can you talk a little bit about what you're seeing so far in the outside environment as it relates to the clearance of last year's models, as well as how pricing is behaving I guess just broadly?

And then I had a follow-up for that.

It's pretty early.

You know, obviously there are lots of clearance sales going on; just pick up any newspaper.

Everybody is overloaded -- let me take that back.

All domestic manufacturers are heavily overloaded with '04 models.

You've heard some of the publicly traded new car dealers talk about cutting back on '05 (technical difficulty) whether they will actually do that or not, we will see because, as you know, it's sometimes hard to get your allocation from new car manufacturer for the hot new model if they have any if you don't order what they want.

But, that's always true at this time of year.

Usually, the beginning of September is a period of time when the trends and the dynamics of the fall haven't made themselves obvious yet.

Our experience is new car dealers are typically optimistic in September about selling the inventory that they acquired at higher prices in August.

And they tend -- we tend not to see the wholesale market fall until at the earliest the end of September, and more likely sometime mid-to-late-October.

So I don't think we've seen anything on the wholesale market side that suggests to us that we have any preview of what's going to happen.

So far it looks like every indicator we have from the wholesale index that we track on cars that we buy, which we've been tracking for four or five years now, suggests that this will be a fairly "normal fall", if you will -- that is with a pretty sharp fall off in wholesale prices as we get into the mid-fall bottoming out in December.

But that's just predicting from where we are through mid-September.

And most of the dynamics really will occur over the next 45 days.

Without wanting to dwell too much on the external, can you comment a little bit on your internal performance, your execution, and just talk to the point of standard stores versus the satellite stores in terms of performance, if you can?

Overall, I think we continue to be pretty pleased with our execution.

I think given a difficult demand environment and a difficult credit environment -- as I said, the average credit walking in our door over the last 90 days or so has been lower quality than it normally is -- we think our stores have actually been doing quite a good job of executing against the customer flow that they have, both in the new stores and the older stores.

The older stores, being more mature, tend to outperform the new stores, but the newer stores are consistent with where we expect them to be and, as I said before, much better than typical new stores were we were opening four or five years ago.

In terms of satellite versus standard stores, the message is pretty much the same as it was last time we talked about this.

The standard stores as a group continue to perform cumulatively in line with what we expect, but there's a pretty wide variance across the group from some stores that have over-performed to some stores that have significantly under-performed, whereas the satellite stores as a group continue to perform slightly ahead, and as a group are more consistent, more predictable.

And that's to be expected because they're generally built in markets where we already have a sales trend, and therefore they're are more forecastable.

So accordingly are there any changes or tweaks or other things?

I know that adding Drive is kind of a slight shift in the model?

Are there any other tweaks that are going on beneath the surface, and then I will turn it over?

There are a million little tweaks that go on all the time.

In terms of significant ones, none that I'd report.

Drive is an incremental addition which helps us sells some additional cars to people who otherwise could not have bought from us.

That's one of the ways that we know it's purely incremental.

Obviously we would not accept the lower margins that Drive brings if we weren't highly confident that it was purely incremental business.

But there's not significant other changes that I'd report at this point.

Thanks a lot.

Scott Ciccarelli, RBC Capital Markets.

A couple of questions.

The first is obviously we've seen at least the first beginning of some production cuts from the domestics.

So how quickly could something like that move through the marketplace and start to impact your business, would be number one?

Number two, Austin, you mentioned a difficult credit environment.

Is there any reason to believe that could start impact the credit quality in terms of the credit portfolio you guys have?

And then the last one, or last question is, any change in the thought process regarding new store openings at this stage?

The first question was production cuts, second question was credit quality, and what was the third one?

Third one was store opening expectations.

As far as production cuts, I'm not a professional analyst following the new car guys.

But all of -- most of you guys have analysts who do, and everything I've read from all of them says this is not enough unless we see a significant, perhaps even dramatic, turnaround in the economy that boosts underlying demand.

So I think it's directionally perhaps a good thing, but it doesn't sound like anybody believes it's going to cure the overhang of inventory.

My understanding, if I've read this right, is that inventories for domestic manufacturers right now are the best they've been this year but the worst they've been at this time of year in 10 years.

So there is quite a lot to be done. and to the degree that the economy is sort of bumping along, or not improving significantly, I think most people believe that what's coming is a bigger adjustment in the first quarter of next year after the manufactures' fiscal year is over.

But that really depends on where you think the economy is going.

I wouldn't expect that's going to have a big impact in our third quarter or their fourth quarter, other than what they have to do to get rid of their '04 models may have a significant impact on how fast prices fall this fall, and maybe even how low they go.

But that's what we always expect this quarter.

So we will be hoping that we see some normalization there, but it doesn't sound like there's enough production cuts that really brings the market back to full formality.

As far as credit quality, what I said is the average quality of the average consumer coming in the door is somewhat lower than it has been typically.

The net result of that is that CAF's penetration has fallen somewhat, because CAF's not going to do lower quality loans.

The net result is our nonprime finance companies have actually picked up more of that business.

So their penetration is up;

CAF's penetration is down.

We manage CAF so that CAF's credit risk should not increase in that environment.

As far as new stores, I noted that we're going to open nine this year.

We have not announced for next year, but I don't see any reason to believe that we wouldn't have a normal store opening pattern in the 15 to 20 percent range next year.

We certainly have the stores all in line to do that, and we're planning to do that.

So although it was a pretty miserable quarter in terms of how we felt about it and how the business performed, the business is still very profitable, and more than probable enough to provide us with what we need to open stores.

And we don't see anything that makes us fundamentally question the effectiveness of the model, just a difficult market environment.

Thanks a lot.

Matthew Fassler, Goldman Sachs.

A couple questions Austin.

First of all, I want to follow up on your answer to one of Aram's questions.

You talked about the expectation that the typical volatility of the third quarter you would see the wholesale markets adjust.

Given the disparity in the value proposition that has emerged between the new car market and late model used cars, would you say that there's a greater-than-typical adjustment that the wholesale market needs to make to get competitive again?

Or would the typical seasonal correction do enough to bring these markets back to equilibrium?

I can only tell you what our experience is.

And don't misunderstand -- it's not that the wholesale market hasn't been adjusting this summer.

If you look at June and July, you saw some of the biggest falls (ph) in large-and-medium-SUVs in a short period of time that we've ever seen in the wholesale market.

It's just that the very high level of activity, the high level of inventories, and the creativity coming out of the new car guys in the middle of peak selling season seems to always get a step ahead of the wholesale market is my judgment.

Once again, that is my best assessment of the situation.

What you ask is the $64 question, which is we know that things will tend to get back in balance in the fall.

The question is will they get all the way back in balance and will it take a bigger than usual adjustment?

And my answer would be, I hope so, because I'd rather see them get back in balance.

And I think the actual dynamic of why that happens is that as we go into a season where demand is declining for pretty much the whole country except for California and Florida -- in other words, driven by the fact the weather gets colder and people buy fewer used cars -- typically, dealers get more conservative on what inventory they're willing to hold in October and November, and it goes to the extreme in December.

So as they do that, to the degree the market is out of balance, you would expect that dealers would see that and adjust accordingly in the inventory they're willing to hold, and that in and of itself would bring the market back in balance.

Once again, that depends on dealers correctly understanding where the new car market is going.

But the new car market I think has been aggressive in enough in terms of its pricing now for awhile that dealers are starting to factor that into their behavior.

I believe that should be true.

The question is will we actually see it this fall.

I think the opportunity is there for the market to fall quite a bit more.

That has pros and cons for us.

Obviously it means that whenever you had that is left over from August is less competitive, and you're going to get lower price on it.

It also means that you get more back in line with the market.

We think we're as prepared as anybody can be for that environment, and we will just have to see how it comes out.

A second question, please.

Looking at the seasonality of your gross margin per used vehicle retail; then in 2002 the seasonality was not that consistent; 2003, a little more so.

But in any event, your gross margin dollars per vehicle retail tend to be a bit lower in the second half than in the first half.

Given that it sounds like you took some lumps in Q2 with some clearance activity, in August would you still expect that gross profit dollars per vehicle to come down in the second half?

Yes, and it's really the third quarter that's the main driver of that.

The fourth quarter you begin to see margins typically recover somewhat, and by the end of the fourth quarter they've recovered substantially.

But really the third quarter margins -- I can tell you, we have taken margins down, so the initial price we're putting on a car has less margin in it, and we always do that in September.

If you remember, the challenges we had last year, one of the reasons is we think we took margins down a little less than we historically might have because of having a couple of years where it seemed like we performed better.

We've gone back to the margin adjustment we would have made three or four years ago because we think that's most appropriate for current market conditions.

It is true, however, that when we compare back against this summer, if the fall performs consistent like that, the fall to summer comparison will look a little bit better probably, just because the summer wasn't good.

Third and final, if I could.

On the credit side you spoke about expecting gain on loans sold towards the lower end of the range I guess because you don't have a lot of flexibility on the APR.

Are you seeing similarly lower cost of funds out there in the marketplace, that that could potentially offset that?

I will let Keith answer.

(technical difficulty) rose in the spring and early summer, but basically have been stable for the last few months, and that's currently what we're anticipating, is that they will remain stable and we don't expect a big change in one direction or another.

Keith, did you say the APR comes down as the (indiscernible) comes down?

In other words, you're getting a bit squeezed -- or did you get a bit squeezed in the summertime, or was it really more of a flattening?

As our cost of funds went up we actually price (indiscernible) with what we saw from an Internet competitive perspective.

And our consumers were telling us through their three-day payoffs that no, they weren't prepared.

And we saw an up-tick in three-day payoffs, so therefore we backed off a little bit.

And that's what our forecast holds for the third quarter, is we expect that the consumer still has low APR expectations given the slow step and slow pace of the Fed increases and an ongoing environment where you still see 0 percent on new cars.

Thanks a lot guys.

Bill Armstrong, CL King & Associates.

Austin, earlier you said that car sales financed by Drive would contribute about 40 percent of other car sales.

Is that net of the discount that you pay to sell off the loans?

Yes.

In other words, net of discounts and commissions, considering all sources of income, they generate about 40 percent as much contribution as a CarMax car would.

So the effect on the P&L from that incremental business would just be additional sales, and obviously cost of sales, and then a negative impact on the other category, right?

That's correct.

And overall it would lower our total dollar earnings per car, and probably lower our margin per car.

Once again, margin per car depends a lot on what retail does, and as we have said before, retail moves all over.

But the real thing we look at is dollars per car sold, and that would come down a bit.

But as long as we're confident that they're incremental, it's still absolutely a sensible thing to do.

Would the low single digit sales effect that you expect going forward, do you expect to see that fully ramp up this quarter or will that take a few quarters?

We don't really know.

What we've seen during the period of time we were testing Drive is that Drive is very sensitive to both the local economic environment around the store -- that is how many people are there who need this sort of financing -- and probably more than anything else we do to understanding how to execute and sell to this customer.

And one of the things that's important to recognize is basically these are customers who in our 11 year history we've never been able to sell cars to.

So each store really you have to understand how do you effectively conduct a process where a customer for the first time is really extremely limited in the number of cars they can buy, because Drive is very limited in the exact cars they will finance.

And it takes some learning.

So that's why we're a little bit cautions about forecasting when we will see this.

It could be this quarter; it might not be until the beginning of next quarter.

We will just have to see.

Does this help you get any additional business in situations where there's negative equity or are the two unrelated?

By and large committees are not people -- actually by and large is the wrong way to say it.

These are not people who qualify for negative equity.

These are not people who qualify for anything.

These are people who every finance source who uses credit ratings has rejected.

So these are people who have fallen out the bottom of the credit rating system, and nobody gives negative equity to those folks.

These folks all have positive equity.

That's a requirement of the loan.

One final question, if I could.

How would you characterize the incentive environment this year versus what we've seen basically since 9/11/01?

We've seen manufacturers be pretty aggressive with incentives for three years now.

And until this year CarMax had very strong sales gains, and you seem to have run into obviously a more difficult environment this year.

How would you characterize this year as being different than the prior two years?

We've done the best that we can to try to understand all the different moving parts.

And there's no question, I think part of what is going on is the overall economy and its direct effect on used car sales.

But I believe -- and I may be wrong, but I believe -- that the bigger impact is the economy's effect on domestics new car retailers.

And in particular I think what has been unusual this year is that the expectations for the late spring and summer selling season by the manufacturers clearly were out of line with what they ended up seeing, and they ended up building a much higher inventory bubble than they expected.

And part of that was driven by the spike in gas prices which had a really negative impact on medium and large SUVs and pickups.

So what we've seen is manufacturers with more inventory in the summer, more days of inventory in the summer than they typically have, and that inventory heavily concentrated in those products in which historically have been most competitive and most profitable against the Japanese.

So really, a very dramatic competitive situation without a lot of certainty on their part as to where the economy is going.

And I think we would described it as -- I don't know that I would go so far as to say chaos, but certainly a very volatile incentive environment right in the middle of the peak selling season.

And that's unusual.

When we look back, we can identify during the Explorer rollover crisis there was a month when Explorers went through very aggressive incentive increases in the middle of the summer.

Once again, it's not the level; it's the change.

But this summer, I think in terms of the incentive changes that we saw especially on large SUVs, small SUVs and pickup trucks, and then later on across many of the cars from the domestic manufacturers, has been more volatile and unpredictable, I think, than anything we've seen.

Logic would suggest that they have to bring their production in line with a reasonable expectation for sales or they will do enormous damage to their residual values, to their dealers, etc.

But in a competitive price war of this magnitude with these stakes, it's kind of hard to predict how the big players are going to play it out.

Is it feasible for CarMax to stock more non-US, late model used cars as a percentage of your mix in order to mitigate that somewhat?

The reality is that everybody is operating in the same market, and the market is pretty volatile.

And we're continuously trying to adjust the inventory at each store to what appears to be the best set of offers we have from the market in terms of what we expect to sell and what prices we can get.

And if we face a better price position and a better sales expectation on non-US models, that's fine.

Don't get me wrong -- when Suburbans plummet enough in price, we can sell Suburbans very nicely; it's just that the price has to come down to the point that somebody's ready to say okay, at that price I will buy one against what the new car guy is offering, and I will buy one regardless of the fact that it only gets 12 miles a gallon.

Our experience is there's always some price at which people will buy the cars.

So just moving away from domestic cars isn't really a good solution.

The issue is trying to manage through this volatility with the best mix that we can get and trying to correctly understand where the consumer is going in terms of what they're demanding.

And we really do that store by store, week by week.

Got it.

Thank you.

Sharon Zackfia, William Blair.

I had some clarifying questions on the relationship with Drive.

The 40 percent net contribution, is that versus a CAF car or versus a blended CarMax car?

Versus a blended CarMax car, and it's after all commissions.

And then I think --

Meaning to our salespeople.

I think you might have mentioned that it might affect the gross profit per car, which I hadn't anticipated.

I thought that would all go through the third party lending line maybe as an offset.

Did I misunderstand what you said?

Yes, I'm sorry.

It will affect -- I used the term gross profit when I mean total contribution.

Total contribution means profit from all sources.

As you know, although it's not how is reported on the P&L, when we look at the business one of the things we look at is what do a make on the car, what do we get from our ACR recovery on wholesale, what do we get from the various finance sources, etc.

And I was speaking in those terms.

Curious, when you're testing Drive, given that this is a new customer to you, did you see any enhanced volatility in your sales trends because this is a perhaps more economically sensitive customer base versus your kind of placebo affect normal stores?

Or how does that trend?

No, not really.

No, because there's a big base of people who would like to be able to use Drive.

Part of Drive's success is they are pretty effective at narrowing to a base of customers and cars that work economically.

One of the reasons we tested for a long time is we don't want to bring on board a subprime supplier that we don't think has a good understanding of their business or a fundamentally sound economic model.

And so, what I'm saying is Drive has a plethora of customers from which to choose, and we haven't really seen anything that adds volatility there.

There's a big base to choose from.

It's really more the store -- the specific demographics around the store, and then the store learning how to effectively execute presenting the Drive offer to a customer, which is very different than the typical CarMax offer.

And then can you remind me, SUVs and pickup trucks and all of those kind of gas-guzzling cars, what percent of your mix do those normally represent?

And where those all meaningfully softer than your reported comps during the summer?

I think the SUVs in total and pickups are in the 25ish to 28ish range, and they were somewhat softer.

Although, during the middle of the summer, actually because of the big fall off in wholesale, there was a period during the middle of the summer where actually SUVs recovered and showed new strength when the prices really went down strongly.

So it really ebbs and flows with where has the wholesale market gotten to versus where the new car retail market is.

You had to leave this door open, so I am going to ask.

The comment on your trends thus far in September being somewhat different than your comp guidance for the November quarter, is that a meaningful difference thus far quarter to date?

I don't recall you having used this language in the past.

You said you're studying your expectations based on the prior quarter's trends.

But I understand why you're doing it.

I'm just curious as to --

The answer is the first half of September is meaningfully better than the summer.

But it's 15 days versus 5 months.

We're happy for those 15 days.

And particularly given that we also had a hurricane, that's pretty good.

Don't get me wrong -- don't get overexcited about that, I'.

Just saying it's better.

But we've got a slower trend for five months, and particularly July and August were quite slow.

So the only thing I'm reporting is that we're pleased with where September started out.

We think there's enough there that it's worth noting to you, but not enough that we think we would want to forecast off of it.

Maybe phrasing it a different way, is this kind of the most encouraging couple of weeks you have had in the last five months?

I don't know.

There been a lot of two-week periods in the last five months.

But I would put it this way -- the first two weeks are consistent with the idea that maybe the market will adjust.

But as I answered to the first question, we're way early in the fall adjustment process, so there's, as they say, miles to go before we're home.

What I will just say is we're happy that the first couple of weeks in September were better.

Thanks so much.

David Campbell, Davenport.

I was wondering if you could talk about Florida first.

Is there any way to quantify the affect of the storm on the business in the second quarter?

And where there any actual inventory losses because of the storms in Florida?

We could quantify it, but it's small.

Let me put it this way -- you wouldn't feel any better.

It definitely had an impact, but we feel equally badly storm or no storm.

It just made a bad quarter a little bit worse.

As far as inventory, nothing of consequence.

We had a port-a-potty blow off a construction site and hit two cars in Orlando if that counts.

But really we were quite lucky.

Frankly, I think the store at the beginning of September had a bigger impact than the one in August.

But because it is at the beginning of September, the chance is pretty substantial that we should recover that if they don't get a bunch more.

So net-net we would hope that that will have washed out by the end of the quarter.

The wholesale margins continue to be relatively strong.

Can you explain why those have been so high and if there's any difference in your thinking about the wholesale business?

No, it's mostly that the ACR program that we put in place is now, if you will, fully operating in the way that we expected to operate.

And pretty much all of the difference comes from that.

It really doesn't change our view of the business.

What it says is the program we put into place to make sure that we recover 100 percent of the costs that it takes us to run the entire wholesale business and to buy cars is working pretty well.

Lastly, incentives appear that they should be relatively strong, given that dealers and manufacturers are continuing to clear inventory this fall.

How do you think that will impact CarMax's business?

The key is does the wholesale market -- and the wholesale market really means new car dealers, used car departments, independent used car dealers and wholesalers -- do those guys understand where the market is going and do they adjust their purchase prices consistent with where the new car market is actually going to go?

And I'm hopeful that it is becoming more clear, having had a pretty disappointing summer for everybody, both in new and used cars.

But one of the reasons that we adjust at the beginning of September is because we know the rest of the market adjusts a little more unpredictably during the next 45 days.

And really I think there's enough information out there that, gee, this is going to be a strong discount season; you ought to be pretty conservative in your offers; you're going to need a good spread to be able to sell these used cars; demand may be somewhat softer; you may want to carry a little less inventory.

But what we will have to see is do people react to those signals and what's the net result.

Thanks.

Good luck.

Charles Grom, J.P. Morgan.

Inventory was up about 20 percent year-over-year for the second consecutive quarter.

What was your days supply?

And could you comment on your mix in terms of Big Three versus imports?

I will let Keith answer that.

First of all, the main reason inventory was up was really kind of twofold.

One, obviously we have seven more stores than we had last year, and we have built six since the beginning of the year.

The other element is you have to consider the timing of Labor Day.

Last year Labor Day included the largest Saturday and Sunday of the month in August -- it ended on a Sunday.

This year Labor Day coming later, we still have an extra inventory level to support Labor Day sales all the way through that weekend.

And then we target actually after Labor Day to be our lowest sales point.

So really the balance of the inventory shift was the combination of those two events -- new stores and the timing of Labor Day.

I don't happen to have the days supply with me, I don't believe.

And the mix between Big Three and imports?

Typically I think we're slightly above 50 percent one way or the other.

I don't know exactly where we are right now, but I don't think it's significantly different than what it typically is.

It might be slightly more import.

But to be honest, I haven't looked at it in the last couple of weeks.

Second question -- new store productivity rate is down pretty significantly year-over-year.

Could you comment on what has led to this decline and whether it's being driven by weakness in your satellite stores or both your satellite and standard stores?

I think the first, most important thing is it's led by incorrect expectations.

As we said some time ago, the stores that we are opening have a projected mature sales rate that is 20 percent lower than the group of stores that we were operating before we started the opening program.

So that was in our sales release a couple of quarters ago.

And I think a lot of people may have missed that.

And there are two reasons for that.

One is because we are opening a significantly higher mix of satellite stores, if you go back to where we were when we started working our growth program with I believe 33 stores, we only had 4 satellite stores in the whole United States out of 33.

We are opening roughly a 40 to 50 percent mix of satellite stores right now.

And satellite stores normally are expected to do less volume because most satellite stores are fill in trade areas, and they are riding on the incremental economics of the fact that you don't have to add much advertising at all to fill in that trade area.

So they are very profitable, but they achieve their profitability at a lower sales level.

And part of how they also achieve their return is they get to that sales level faster.

The second reason that the expectation is about 20 percent lower is that we are opening more small mid-size markets -- in other words, more Columbia, South Carolina's and Knoxville, Tennessee's -- when historically we didn't have any markets that small in our base.

And we are opening them because everything we've learned suggests that those are -- while they're lower sales volume, they are somewhat more predictable than the bigger markets, particularly if they are close to an established CarMax trade area and have a base of CarMax customers in the market who already know us so that we have word of mouth, return customer, etc.

So most of the difference -- in fact, almost all of the difference -- is accounted for by the fact that we expect these stores to do lower volumes, which doesn't mean that the economics aren't attractive.

In fact, the economics are very attractive.

Having said all of that, the last five months have been a lousy five months for the entire business, new stores and old stores.

On average I think new stores have actually probably done a little better than expectation than the old stores have.

But I'm not sure how meaningful that is because the old stores have a lot of comp history to compare to, whereas the new stores just have the budget that we set for them.

So overall in terms of performance to expectation, the new stores are in line with what we've expected of them, although everybody's suffered some softness this summer.

And the lower sales per store is more driven by the mix of stores we are opening.

And a natural follow-on question might become be well gee, do you think you will at some point in the future open more large stores?

And as we go into larger multi-store markets, there's certainly more opportunity to open more of the large stores like we have in the Washington/Baltimore area, in the Atlanta area, etc.

Obviously those are more complicated and riskier markets.

Typically over the long-term you do very well in them, but often over the short-term they're less predictable, harder to get real estate in, harder to get ramped up initially.

Or you get an incomplete set of stores and you have, therefore, more advertising load per car sold.

That's why we have said we would wait for three or four years into the opening program before we started opening stores like that.

And you can really look at Los Angeles and say that is the first of those markets that we're taking on, not necessarily because it would be logical to start with LA, which is one of the two biggest markets in the country, but really because we were sort of half pregnant (ph) there when we stopped our growth program and already had two stores open and two additional pieces of real estate, so that we were committed enough that we think that makes it the next most logical market.

But as we go into bigger markets like that we will open some stores that have higher sales expectations.

I will tell you that just like the larger mid-size markets, we will expect a higher dispersion in terms of some will perform better and some will perform worse when we do that.

It that helpful?

Extremely helpful and you did answer my follow-up question.

But Just one more.

You opened five stores last year; you opened nine this year.

Is the expectation to open 8 to 10 next year or is it more on a 10 to 12 kind of level?

We have not given it yet.

What we have said is we will be in that 15 to 20 percent range.

Thanks a lot and good luck.

Michael Novak, Frontier Capital.

My first question is you had mentioned in the press release that the data you were looking at suggests that you are maintaining or gaining market share.

Could you share specifically what data that is that you're looking at?

We look at two things.

Well, we look at three things.

We look at the reported sales of the publicly traded new car dealers, and we adjust those -- their quarters and our quarters don't match, but the good news is we know what our monthly comps are, so we adjust those and compare apples to apples.

And up through the point that they've announced, which is through the first six months of the year, of the calendar year, we in fact continue to run at a significant spread against them -- not a double-digit spread, but a mid-single-digit spread.

The other two sources are the two main sources of DMV data -- Auto Counts (ph) and Cross Sell (ph) -- and we buy both of them.

They both collect data in the same markets, they come up with different results, and their results are highly volatile month-to-month.

But we look at those results on a rolling trend, we compare them to each other, etc.

And all of that data suggests that market by market we're either maintaining or gaining share typically.

Some markets we will lose share, but some markets that's always true.

But as a group we think we've at least maintained share and probably gained share some.

But I always take the DMV data with a grain of salt because, as I have said before, we see a 10 to 40 percent error rate when we compare the data they report on us versus actual sales data on a monthly basis.

So it takes quite a while before you get a big enough sample that it starts to even itself out.

My second question, in the past when it's a very difficult sales environments, have you seen following that used dealers and perhaps yourselves decrease your inventory commitment so that there is less demand for the cars, and then the prices fall, and it's itself adjusting?

And if that's the case, why has it been so long to occur this time?

That's a great question.

The answer is, as far as we go we generally align our inventory with sales rate literally every week.

I believe one of the things that has been unusual in the last five months is that it's not typical that you have the sort of situation that we've seen where there was a lot of optimism going into the spring, and then a really disappointing spring and summer selling season where you particularly had almost desperate new car manufacturers in terms of their incentive behavior, and I think incentive behavior that therefore went out and took away some of the upper end of the used car market.

It's a combination of both of the volatility of that behavior -- and that's driven by the fact that the new car guys certainly went into the summer expecting even though the spring wasn't that good surely things will get better.

And surely things never did get better, but everybody was expecting they would.

And all of us during the summer I think had that attitude that surely the summer can't be as lousy as it looks like it's going to be, and surely July is going to be better than June, or August is going to better than July.

And I think the net result was that it wasn't.

And I don't recall ever seeing that situation in the 11 years we've been in the business through the peak spring and summer periods.

And I think that dealers began to adjust inventories somewhat during the summer.

But you're still right in the middle of your peak selling season.

And as I said before, if you bring inventories down and it turns out the sales potential is there, it's almost impossible to recover and build inventories because you're at peak sales level, you don't actually have -- you're also at peak sales service level.

So you don't have the reconditioning capacity to add back 50 cars in a given store if it turns out that there's more demand there than you expected.

I think the general behavior is to be more aggressive on inventory during the summer.

And usually most people's experience is, they can't be too aggressive in the summer.

They turn all the inventory they can stock pretty well.

That certainly was not our experience this summer, and I don't think it was other peoples'.

And that's why I would presume that we'll see a significant adjustment this fall, but I want to see it before I believe it.

It should happen.

Logic says it should happen, but logic doesn't always prevail.

If it wasn't in the case, what would you go to as the structural reason why that didn't happen?

I know you have talked about negative equity in the past, but has there been any structural shift in the market?

Look, I think negative equity is an influence, but if it doesn't -- and I think it's not an issue of whether it's not going to happen;

I think it's an issue of when it's going to happen.

Is it going to happen in 90 days, 6 months, or could this go on for 18 months?

It seems extraordinarily difficult to me that it would go on for 18 months and that used car retailers wouldn't adjust their behavior.

That would be unusual because certainly, for instance, after 9/11 we saw people adjust their behavior very quickly.

Now, everybody knew -- everybody thought they knew where things were going after 9/11, which is everybody thought they were going straight down a rat hole.

So everybody moved in one direction and you could argue that in fact they over-adjusted.

And that was good for us because we didn't.

I think the uncertainty of direction is the real issue here.

And if manufacture behavior is somehow inconsistent with what you would expect in the fall and winter, that's the only reason I can see that dealers wouldn't adjust to the actual environment.

And there is a more consistent expectation, I think, with what you're going to see in the fall and winter.

But if you had asked me five months ago, I would've told you that the pattern we saw April through August was highly unlikely, if not impossible.

So, obviously, I was wrong on that.

So I don't want to say that it's going to change in the fall.

Because the market can always through you some new curve you haven't seen before.

And I guess what I'm saying is, if it changes in some other way, we will be trying to figure out what's going on here that we didn't understand?

Thank you.

Matt Nemer, Thomas Weisel Partners.

Just wondering as you look at your cost structure, given that you expect to be down a little bit in the third quarter, is there anything that you can do -- and I think we've talked about this in the past, but can you ratchet back anything in the SG&A line to sort of counteract that?

Obviously as your sales volume is lower, your commission adjust and your sales force size adjusts.

And that's one big thing.

And also, technician level doesn't adjust as immediately, but in general you're always out hiring technicians.

So if sales are going to be a little lower we will have the opportunity a hire fewer.

But beyond those things, there are not obvious quick changes you can make.

We're going as we always do, but circumstance always provides some more push to look at what are some things that we could do to improve cost efficiency that we haven't thought of yet.

So what I am saying is there are not obvious levers that you could knock down, particularly as you're still growing stores.

We're hiring people, we're moving people, we're training people.

That has a cost base and there's nothing that we've seen so far this says we should stop growth or significantly cut growth back below the rate that we've projected.

It wasn't a particularly happy quarter, but it's still more than a good enough quarter to justify the growth program that we have.

So really it will come down to more can we find some innovations or some areas that we haven't thought of before that will give us a pick up there.

The point of view I would take if I were you is I would not expect them so that you'll be pleasantly surprised if we find them, because if they were obvious we would have already done it, and we would not have waited until difficult times.

But difficult times always stimulate you to be even more aggressive in trying to look for things.

Can you give us an update on overall credit availability in the industry and I guess from your third party providers?

You're talking about to our consumers?

Yes, that's correct.

Overall it's been pretty consistent and pretty good.

I think we've seen a pretty stable environment with both our prime and nonprime suppliers.

As I noted, we've seen a modest down-tick in the percentage of mix of A credit customers, if you will.

But for the customers coming through the door, we've seen a consistent supply for folks who are coming through the door.

We like to see some of those A credit customers back.

And I think as used cars, negative equity and incentives get a little more in line, we would expect eventually to see that normalize and get back to where it was.

But we haven't seen any shortfall in credit offering from any of these guys.

The last question is do you expect to receive any consideration for your other Mitsubishi franchises?

And how many do have left?

Where are we at, three?

We have three right now.

Mitsubishi franchise isn't worth a whole lot right now.

You want to buy one?

Look, if so, it will be modest.

We have had them for sale for over a year.

The particular reasons we were selling Mitsubishi don't actually relate to the challenges that they've gotten into as a company.

But the timing will probably cost us a little bit of money per franchise.

But actually, whether we had sold at the peak or whether you sell at the bottom, it's not probably material on any one of them.

They never did sell for very much, and now they're selling for a little less than that, if at all.

Thanks very much.

By the way, we did not pay for several of them.

They were new point grants (ph).

Thank you.

Stephen Eisen (ph), FrontPoint.

Thanks for taking my question.

Just a question on the new subprime relationship.

Just in looking at companies -- customers that do buy those type of -- have those type of credits, particularly they buy based on the monthly payment rather than the absolute dollar price of the car, and you're famous for really selling at the best price.

I am just wondering how prepared is the sales force to deal with something that had been may be slightly different in terms of selling the car to a different type of customer?

The reality is a lot of our customers who buy from nonprime or even some of the ones who buy from prime are monthly payment focused.

You have a whole mix of customers coming in the door.

The issue is more because we do sell at fixed prices, obviously what we don't do is what another dealer might do, which is sell at a higher price to a customer who is using a subprime source than some other source.

A typical dealer might pay a high discount on subprime, but not come at all off his very high initial starting price.

And that's obviously not our behavior.

And we haven't changed behavior in the store with that regard.

So it's more this is a customer who has to go through a whole different -- goes through our entire application process, gets turned down by everybody else and has to go through an entirely different application to drive; and has to have a certain amount of money down; is only allowed to select from a certain limited number of cars because that's all Drive will finance.

So that's what's really different about the selling process, not the fact that it's a monthly payment customer.

Thank you.

Mike Heifler, Deutsche Bank.

Thanks for taking my question.

Austin, I just want to follow up on the earlier market share question and the competitive landscape.

I'd like to get your thoughts on the certified used programs.

The manufacturers seem to be stepping up in this area, and they're even offering subvented rates at this point.

Could this become an issue for you guys as rates go up?

Look, that's been out there for a while actually.

Subvented rates have been coming and going.

Clearly I think in theory all manufacturers would love to have a high level of certified sales because they feel like basically if you can sell a certified car you're selling a car a higher value, and you may be protecting the brand better.

What you've seen if you've watched over the last couple of years is the rate of increase of certified sales has slowed dramatically to just barely any increase at all.

And most of the increase recently has been coming from people who are new to certified programs or relatively smaller.

The most successful programs, Toyota and Honda, have really flattened out, even declined a little bit.

And the other most successful programs are Mercedes and BMW.

And a consistency that you will find across all of those it is the stronger your brand value, the more your brand means, the better chance you have of convincing a consumer to pay a premium for a car in order to get the manufacturer's stamp of approval.

The manufacturer doesn't do anything on the car other than provide the dealer with a powertrain warranty.

But dealers do typically charge more for certified cars and manufacturers have some standards.

In fact, the Chrysler and Toyota programs were pretty much exactly modeled after the CarMax program.

So our experience is, it's another factor out there that for better car dealers helps them have one more arrow in their quiver to say, look, we are as good is CarMax.

It doesn't change their pricing behavior or the integrity with which they deal with the customer.

That was either good or bad, whatever it was at the particular dealer.

But it does mean they have a better case to make on the quality of the car.

Our experience is that in general they charge enough for that that we generally have the same quality certification at a lower price.

Then the question often depends are you so much of a Toyota fanatic that you really would prefer to buy something that has the manufacturer's stamp of approval?

And there's some people who would, some who wouldn't.

In fact, we sell both types side-by-side in Laurel, Maryland -- we sell Toyota certified cars at our Toyota store; we sell CarMax certified Camrys next door.

CarMax certified Camrys are typically a better deal because they don't have as expensive -- they don't have the manufacturer charge.

They just have the cost of what we do.

Somber prefer one, some prefer the other.

Our net conclusion is it does not change the competitive terms meaningfully.

It is not to say that if you're a dealer you might not want to do it.

At the margin it's not a bad thing for you, but it has a cost.

We don't believe it's changing the competitive terms substantially.

What we've seen from the subvented financing is that it's not a big enough amount or impact that we it fundamentally changes any sort of long-term competitive behavior.

On an individual vehicle, for a short period of time it can have an impact, but broadly it doesn't seem to.

Thank you.

I think this is the last question.

Hardy Bowen (ph), Arnhold & Bleichroeder.

I guess I just didn't get in time to get at the front of the queue.

The SG&A seemed to come out lower than what we would have projected.

And I think adjusting for satellite stores it is about 20 percent increase in average units quarter-to-quarter in the second quarter.

Is that happening because of small markets like Columbia and Greensboro that is doing that?

Or is something else at work?

Keith is chuckling.

It's not really very funny.

We relieved the management bonus during the quarter.

Excellent.

All of us in the room.

So we made our contribution, and that's because consistent with bonus program we didn't earn it this summer, so we didn't accrue it.

Do we have the feeling that marketing we should be approaching things differently in this kind of environment at all or not particularly?

I guess there's so much noise out there that you lose some traction as far as gaining customers, the long-term trend of all these stores in this kind of environment.

But I don't know if there's anything to do about that with advertising.

Do you have any thoughts?

Lots of thoughts.

I've been driving our marketing guy crazy with my thoughts all summer.

And we have run a fair number of experiments.

By and large we have run -- stepped-up television advertising, added radio advertising, and substituted more TV and radio for newspaper run, more aggressive Kelly Blue Book comparison advertising, taken some more price-focused adds that we hadn't used in awhile in put those back on TV.

We've done a fair number of things, and it's as frustrating as advertising usually is.

It's tough to do something that moves the needle enough that you can tell you did anything.

Every once in a while you stumble across something like that, it's a real victory.

But unfortunately I have to report that for all the experiments we did this summer, we didn't find any thing that moved the needle enough to say, okay, we want to change and follow that course.

That doesn't mean we will quit doing them, because if you don't keep trying you'll never find it.

And over the longer term there are some insights that I think we have had recently that might have some longer-term benefit.

But we need to do more testing on those, and they really don't have anything to do with current market environment, just a couple things that I think we're learning in marketing that might be the next increment of benefit.

So it's just a high rate of experimentation to find things out?

And what you have to do is balance.

Pretty much any marketing department can spend all the money you will give them on experiments.

So what you have to do is figure out how many can you afford to run, and how can you structure them in the best possible way so that if something is happening you can detect it.

And we've tried to do that as prudently as we could through the summer, but make sure that we have a pretty high rate of them.

And we will be doing more of it in the fall.

I guess Drive is kind of the premium end of the pay-as-you-go business almost.

Do you plan to market it to pay-as-you-go customers to try to attract them?

No.

Our experience is we have plenty of customers coming in the door who are already eligible for this, and the first thing we need to do is to understand how to effectively offer it to the customer flow that we already have.

And I don't think we feel like we need to change our marketing in any way.

We've always driven in lots of hopeful customers who turn out just not to have the financial capability of buying the quality and price of car that we have.

And what this does is it means that some percentage of those hopeful customers will in fact be able to buy from us.

So you want to convert those customers first and find out how to do it?

It's a very -- before you go out and spend money on a very, very small segment, just look at the economics.

It's not worth spending much money on.

If you can get it as a free increment off what you have, that's one thing.

But to go spend any dollars per car to advertise to get these folks in when they already earn us much less per car that anybody else, wouldn't be worth a lot.

Right.

Okay, sounds good.

Thanks very much for joining us.

We appreciate it.

This conclude today's conference call.

You may now disconnect.